Diet, Diabetes, and Serum Lipids

Although virtually every textbook of medicine recommends specific dietary treatment for diabetes mellitus and every diabetic clinic counts its dietetic department of prime importance in patient therapy, the disquieting fact remains what is the proper diet for the diabetic patients is still questionable. The exploration of various possibilities, so active in the pre-insulin days, came to a near standstill for many years after the discovery of insulin. With the recent realization that the complications of diabetes, far from being prevented by insulin, are being seen with greater frequency has come a resurgence of interest in dietary treatment of diabetes. The emphasis is now different, however. In the pre-insulin era, diet was designed to prevent death from diabetic acidosis. Now the prime aim of diet is to prevent the vascular complications of diabetes. Vascular complications of diabetes may be divided into two groups which are in all likelihood of different etiology. Small vessel disease, such as seen in diabetic retinopathy, nephropathy and possibly in neuropathy, appears to be specific for diabetes. Large vessel disease as in the coronary artery or cerebral arteries is atherosclerotic in nature and while more common in diabetics than nondiabetics is not different from that of nondiabetics. Our observations concerning diet, serum lipids and complications of diabetes in the 1950's compared with the 1930's, together with other studies in the broader field of lipid metabolism and vascular disease, suggest certain premises which are speculated on herein. 1. Small vessel disease does not appear to be influenced by diet or to be associated with any particular serum lipid pattern. The prevalence of retinopathy and neuropathy in New Haven was similar in the 1930's when high fat diets (65 per cent of calories) were employed, in the 1950's when average fat diets were used (40 per cent of calories), and in Japan where poor control, frequent marked glycosuria and very low fat diets (10 per cent of calories) are the rule. The Japanese study disclosed milder retinopathy and rare diabetic glomerulosclerosis, possibly related to the preponderance of early diabetes in that series. The high serum lipids in the New Haven study compared with the low lipids in the Japanese study provide further evidence of the lack of correlation between serum lipids and small vessel disease. 2. The atherosclerotic complications of diabetes are related to serum triglyceride concentration. With increased atherosclerosis of diabetics there has been observed an increase in the serum triglycerides. Furthermore, atherosclerotic disease in diabetics as in nondiabetics is associated with elevated serum triglycerides. We have found this relationship to be more significant than the relationship between cholesterol and atherosclerosis. A dissenting note is that of New et al. who found triglycerides were elevated in diabetics with coronary artery disease between the ages of thirty-five and fifty but not thereafter. A re-examination of our data, considering only males because we have a large series of normals for comparison, revealed that 82 per cent of seventeen male diabetics with clinical atherosclerosis over age fifty have elevated triglycerides, compared with 50 per cent of twelve diabetics of this age group without complications and 38 per cent of 159 normal men over age fifty. Both studies of diabetics are too small to expect agreement and clearly more study is needed. In a large series of diabetics Schrade and colleagues also reported elevated triglycerides in older atherosclerotic diabetics. Whether any particular class of triglyceride rich particles in the circulation or all abnormal particles are implicated' needs further clarification but in the meantime elevation of triglycerides above 5.5 mEq./L. (about 140 mg. per cent) is a simple albeit rough indicator of the presence of such particles. 3. Serum cholesterol concentration is not related to vascular disease in diabetes. Although cholesterol concentration differs greatly between certain populations such as the North American and the Japanese, the prevalence of small vessel disease in Japan was similar to that in the United States. We found no association between serum cholesterol and either small or large vessel disease. Nonetheless since diabetics have slightly higher serum cholesterol concentration than normal people the effect of diet on cholesterol must be considered.-When the effects of weight loss are discounted, most efforts to lower cholesterol result in only a small reduction.